基于双目相机的车辆鸣笛监测方法

Vehicle Whistle Monitoring Method Based on Binocular Camera

为了减轻交通噪声对日常生活的影响,对车辆违规鸣笛行为进行监测是必要的。提出了一种结合双目相机的车辆鸣笛监测方法,通过对双目相机采集的深度图像可计算出定位区域与麦克风阵列间的距离,对深度图像进行分割获得有效定位区域以减少定位算法计算量,提高违规鸣笛监测声源定位的效率与准确性。通过车辆鸣笛定位实验,验证了该方法能够准确进行鸣笛车辆定位,相较于传统车辆鸣笛监测方法减小了82%定位计算量,提高了鸣笛监测计算的效率。

In order to reduce the impact of traffic noise on daily life,it is necessary to monitor the illegal honking behavior of vehicles.A vehicle whistle monitoring method combined with binocular camera is proposed.The distance between the location area and microphone array can be calculated by the depth image collected by binocular camera.The depth image is segmented to obtain effective location area to reduce the calculation of location algorithm.So as to improve the efficiency and accuracy of sound source location for violation whistle monitoring.The results show that the proposed method can accurately locate the whistle vehicle,which reduces the calculation amount by 82%compared with the traditional monitoring method,and improves the calculation efficiency of the honking monitoring method.